loandbehold

6 ms·

31.
by loandbehold 10mo ago
If you like I can try to help you getting Claude Code to handle your incredibly poorly written program. loandbehold0 at proton.me
32.
by loandbehold 10mo ago
Email me loandbehold0 at proton.me with details of what you are trying to do. I don't have quantitative finance experience but I can give it a shot.
33.
by loandbehold 10mo ago
You can opt-out of having your code being trained on. When Claude Code first came out Anthropic wasn't using CC sessions for training. They started training on it starting from Claude Code 2 that came out with Sonnet 4.5. User is asked
34.
by loandbehold 10mo ago
The reason is because we are a Microsoft shop and our company doesn't have Claude account. I'm using my personal Claude Max account. My manager does know that I use Claude Code and I requested the person responsible for AI tooling
35.
by loandbehold 10mo ago
Senior Software Engineer. The system is a niche business software software for a specific industry. It doesn't do any fancy math, all straightforward business logic. > Trying to get clean code erases all/most of my productivity
36.
by loandbehold 10mo ago
For me, as a remote developer, it means I'm able to finish my work in 1 hour instead of 8 hours. So I'm able to capture "extra value" in the form of time. In our team everyone uses GitHub Copilot and I use Claude Code. M
37.
by loandbehold 10mo ago
Most of software work is maintaining "legacy" code, that is older systems that have been around for a long time and get a lot of use. I find Claude Code in particular is great at grokking old code bases and making changes to it. I
38.
by loandbehold 10mo ago
Charging big battery takes the same amount of time as charging small battery if charger's power is proportionally larger. E.g. charging Tesla at Supercharger takes same amount of time as charging phone using fastest phone charger.
39.
by loandbehold 10mo ago
Doesn't work on mobile browsers.
40.
by loandbehold 1y ago
I keep hearing that LLMs are trained on "Internet crap" but is it true? For instance we know from Anthropic copyright case that they scanned millions of books to make a training set. They certainly use Internet content for trainin
41.
by loandbehold 1y ago
Because training frontier model is expensive and only US and China have capital structure to raise tens of billions of dollars to do it.
42.
by loandbehold 1y ago
Aren't all frontier models already able to use all these languages? Support for specific languages doesn't need to be built in, LLMs support all languages because they are trained on multilingual data.
43.
by loandbehold 1y ago
People like Ray Kurzweil, Jürgen Schmidhuber, Leopold Aschenbrenner, Paul Christiano see a pathway out of this inevitability through technological singularity. In their vision humans are a seed to something greater, something noticeable on
44.
by loandbehold 1y ago
That's where glyphosate comes in handy.
45.
Behind Russian Drone Surge? Chinese Factories(washingtonpost.com)
7 points by loandbehold 1y ago|0 comments
46.
by loandbehold 1y ago
That's an example of incompetence, not Dunning-Kruger.
47.
by loandbehold 1y ago
What does this have to do with Dunning-Kruger?
48.
by loandbehold 1y ago
The US is second by manufacturing output after China. It's just not true that "US produces very little".
49.
China's Missing Boasss(economist.com)
2 points by loandbehold 1y ago|1 comments
50.
by loandbehold 1y ago
Why do you need an agent use web app through UI? Can't agent be integrated into web app natively? IMO for verticals you mentioned the missing piece is for an agent to be able to make phone calls.
51.
by loandbehold 1y ago
Users' chat history is the moat. The more you use it, the more it knows about you and can help you in ways that are customized to particular user. That makes it sticky, more so than web search. Also brand recognition, ChatGPT is the de
52.
by loandbehold 1y ago
Whenever I see Claude Code make same mistake multiple times I add instructions to clade.md to avoid it in the future.
53.
by loandbehold 1y ago
It's quite common for some laws to be broken on a regular basis to the point where it becomes expected behavior. Going 70 mph when speed limit is 65 mph is one example.
54.
by loandbehold 1y ago
Why not other way around?
55.
by loandbehold 1y ago
Both. But I'm not talking about fixing. I'm talking about refactoring.
56.
by loandbehold 1y ago
I read about AI assistants allegedly creating tech debt but my experience is opposite. Claude Code makes it easy to refactor helping to reduce tech debt. Tech debt usually happens because refactoring takes time but is hard to justify to upp
57.
by loandbehold 1y ago
How does it kill open source products? The only thing required is for open source contributor who is responsible for publishing .apk to present their ID.
58.
by loandbehold 1y ago
You are free publish source code for your app. You are free to publish unsigned .apk and people who want it will find a way to install it. Once app is installed it's more than a speech: it's a potentially hazardous product. The an
59.
by loandbehold 1y ago
I agree, people here have no empathy to less technical users. Peoples' lives being ruined is not a hyperbole. You have people losing their life savings due to pig-butchering scams and such. And people here think their convenience and d
60.
The Tale of the Rent-Seeking Saw(wsj.com)
4 points by loandbehold 1y ago|2 comments

More ›